teknium's picture
Update README.md
c45f845
|
raw
history blame
773 Bytes

Base Model: GPT4-x-Alpaca full fine tune by Chavinlo -> https://huggingface.co/chavinlo/gpt4-x-alpaca
LORA fine tune using the Roleplay Instruct from GPT4 generated dataset -> https://github.com/teknium1/GPTeacher/tree/main/Roleplay
LORA Adapter Only: https://huggingface.co/ZeusLabs/gpt4-x-alpaca-rp-lora/tree/main/gpt-rp-instruct-1

Merged LORA to the model.

Instruct it same way as alpaca / gpt4xalpaca:

### Instruction:
<prompt>
### Response:

or

### Instruction:
<prompt>
### Input:
<specific data to manipulate for the instruction
### Response:

For a better idea of prompting it for roleplay, check out the roleplay discord bot code I made here: https://github.com/teknium1/alpaca-roleplay-discordbot


tags: - llama - alpaca - gpt4